Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down at the end of May, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Nick Cox <njcoxstata@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: r(198) error using lrtest to test for heteroskedasticity. |

Date |
Sat, 7 Jul 2012 01:44:03 +0100 |

You omitted to estimate the model again. Therefore -hetero- is still the most recent set of estimates and you are comparing it with itself. The -lrtest- hinges on comparing two different model fits. If you look again at the FAQ you will see that there are two model fits. (I first looked at your bottom line, which looked wrong. I didn't notice in my first posting that you made two mistakes.) Nick On Sat, Jul 7, 2012 at 12:07 AM, Sean Lim <seanlim174@gmail.com> wrote: > Dear Nick, > > Thank you for your reply. > > I have tried the commands with the alterations you suggested. > > I am afraid I still have the same error even when I include the period(.). > > . lrtest hetero . , df(20) > models hetero specified more than once > r(198); > > Could this be an issue with my dataset? I have a very unbalanced panel > with many data points missing in the independent variables. > > On 6 July 2012 19:24, Nick Cox <njcoxstata@gmail.com> wrote: >> The FAQ recommends different syntax. You omitted the period (.) in >> >> lrtest hetero . , df(`df') >> >> Nick >> >> On Fri, Jul 6, 2012 at 1:15 PM, Sean Lim <seanlim174@gmail.com> wrote: >>> Dear Statalisters, >>> >>> I am trying to test for heteroskedasticity in a panel dataset I have. >>> I am following the steps described in >>> http://www.stata.com/support/faqs/statistics/panel-level-heteroskedasticity-and-autocorrelation/ >>> but I am getting an error: " r(198) error when trying to use lrtest. >>> >>> >>> Any advice on how I should continue with this is appreciated. Below is >>> my output. >>> >>> Many thanks, >>> >>> >>> Sean >>> >>> . xtgls leverage avol7 time_tomaturity grade1 bid_ask_predict, igls >>> panels(heteroskedastic) >>> Iteration 1: tolerance = .03519474 >>> Iteration 2: tolerance = .10619771 >>> Iteration 3: tolerance = .13432693 >>> Iteration 4: tolerance = .12540942 >>> Iteration 5: tolerance = .0969322 >>> Iteration 6: tolerance = .02952538 >>> Iteration 7: tolerance = .01171261 >>> Iteration 8: tolerance = .00374088 >>> Iteration 9: tolerance = .00121751 >>> Iteration 10: tolerance = .00043942 >>> Iteration 11: tolerance = .00016508 >>> Iteration 12: tolerance = .00006333 >>> Iteration 13: tolerance = .00002464 >>> Iteration 14: tolerance = 9.688e-06 >>> Iteration 15: tolerance = 3.844e-06 >>> Iteration 16: tolerance = 1.537e-06 >>> Iteration 17: tolerance = 6.189e-07 >>> Iteration 18: tolerance = 2.506e-07 >>> Iteration 19: tolerance = 1.019e-07 >>> Iteration 20: tolerance = 4.167e-08 >>> >>> >>> Cross-sectional time-series FGLS regression >>> >>> Coefficients: generalized least squares >>> Panels: heteroskedastic >>> Correlation: no autocorrelation >>> >>> Estimated covariances = 21 Number of obs = 15018 >>> Estimated autocorrelations = 0 Number of groups = 21 >>> Estimated coefficients = 5 Obs per group: min = 110 >>> avg = 715.1429 >>> max = 782 >>> Wald chi2(4) = 953523.91 >>> Log likelihood = 13896.44 Prob > chi2 = 0.0000 >>> >>> ------------------------------------------------------------------------------ >>> leverage | Coef. Std. Err. z P>|z| [95% Conf. Interval] >>> -------------+---------------------------------------------------------------- >>> avol7 | -.7824862 .0049243 -158.90 0.000 -.7921376 -.7728347 >>> time_tomat~y | -.0002712 .0000154 -17.65 0.000 -.0003013 -.0002411 >>> grade1 | -.399934 .0010845 -368.76 0.000 -.4020597 -.3978084 >>> bid_ask_p~ct | .0661513 .0043105 15.35 0.000 .0577029 .0745996 >>> _cons | .9815676 .0009443 1039.42 0.000 .9797167 .9834184 >>> ------------------------------------------------------------------------------ >>> >>> . estimates store hetero >>> >>> . local df = e(N_g) - 1 >>> >>> . di `df' >>> 20 >>> >>> . lrtest hetero , df(20) >>> models hetero specified more than once >>> r(198); >> * >> * For searches and help try: >> * http://www.stata.com/help.cgi?search >> * http://www.stata.com/support/statalist/faq >> * http://www.ats.ucla.edu/stat/stata/ > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/statalist/faq > * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**Re: st: r(198) error using lrtest to test for heteroskedasticity.***From:*Sean Lim <seanlim174@gmail.com>

**References**:**st: r(198) error using lrtest to test for heteroskedasticity.***From:*Sean Lim <seanlim174@gmail.com>

**Re: st: r(198) error using lrtest to test for heteroskedasticity.***From:*Nick Cox <njcoxstata@gmail.com>

**Re: st: r(198) error using lrtest to test for heteroskedasticity.***From:*Sean Lim <seanlim174@gmail.com>

- Prev by Date:
**Re: st: r(198) error using lrtest to test for heteroskedasticity.** - Next by Date:
**Re: st: r(198) error using lrtest to test for heteroskedasticity.** - Previous by thread:
**Re: st: r(198) error using lrtest to test for heteroskedasticity.** - Next by thread:
**Re: st: r(198) error using lrtest to test for heteroskedasticity.** - Index(es):